Mental Health Advertising: A Framework for Admissions

Segmenting by modality: medication-seeking vs. therapy-seeking

Treatment demand does not arrive as a single audience. The 2024 CDC National Health Interview Survey brief reports that 19.3% of U.S. adults took medication for their mental health during the prior 12 months, while 14.0% received counseling or therapy from a mental-health professional in the same period ¹. That gap is not a rounding error. It represents two overlapping but distinct populations whose search language, referral pathways, and evaluation criteria diverge in ways that generic brand campaigns flatten.

Medication-seeking adults tend to enter through psychiatric prescribers, primary care, or telepsychiatry channels. Their queries skew toward specific diagnoses, medication names, side-effect concerns, and prescriber availability. Landing pages that emphasize psychiatric evaluation timelines, medication management protocols, insurance acceptance for prescriber visits, and refill logistics convert this cohort. Copy that leads with talk-therapy modalities does not.

Therapy-seeking adults filter on modality (CBT, DBT, EMDR, trauma-focused), clinician credentials, session format, and cadence. Their decision windows are longer and their objections concentrate on therapist fit and continuity. A single generic “mental health services” landing page collapses both audiences into one and typically underperforms segmented pages on cost per qualified inquiry.

The operational move is to build service-line pages and ad groups that mirror the modality split rather than the org chart. Medication management, outpatient therapy, intensive outpatient, partial hospitalization, and residential each need dedicated creative, headlines, and intake questions. Reach calibration then becomes measurable: impression share by service line, qualified inquiry rate by modality, and admissions weighted against the demand mix present in the local market.

Decision moments across search, social, and referral

Reach also fragments by decision moment. A prospective patient researching a diagnosis at 11 p.m. is not the same audience as a family member calling on behalf of an adult child mid-crisis, and neither behaves like a discharge planner routing a referral from a hospital partner. Each moment carries its own channel, latency, and evidence burden.

Branded and non-branded search capture active-intent inquiries but sit in the FTC’s line of sight after the 2025 Evoke Wellness settlement, which turned on Google search ads that impersonated other treatment providers ⁵. Non-branded queries — city plus diagnosis, modality plus insurance — need landing pages that clearly identify the operating provider, the physical location, and the clinical scope. Ambiguity here is now a documented liability, not a creative choice.

Social platforms serve earlier decision moments: symptom research, stigma reduction, and family influence. Reach targets attitude change and consideration rather than same-day inquiry. Referral channels — primary care, EAP networks, hospital discharge, drug courts — operate on a professional-to-professional trust cycle measured in months.

A defensible reach plan assigns each channel a role in the pathway and a specific expected outcome. Search carries active inquiries into assessment. Social carries awareness into brand recall for later search. Referral carries clinical trust into predictable weekly admissions. Blending the metrics — treating a social impression as equivalent to a paid-search click — is what produces the dashboards that look healthy while census stays flat.

The vocabulary that draws FTC attention

Certain words carry documented enforcement history. Marketing leaders can treat them as automatic triggers for clinical and legal review before any ad, landing page, or call script ships. The pattern is visible across three cases in three years.

In 2022, the FTC obtained a $3.8 million civil-penalty judgment against R360 over alleged misrepresentations that its referral network handpicked and evaluated treatment centers through individualized clinical criteria ⁸. Words like handpicked, elite, vetted, and matched imply a substantiated selection process. If the underlying process is a paid-placement arrangement, the language is a material claim the operator cannot back.

In 2023, the FTC pursued a medical clinic over unsupported claims that its services could cure addiction, resulting in a $100,000 civil penalty and an order barring the claims ⁹. Cure, guaranteed, permanent, rapid recovery, and any specific success-rate figure fall in the same category. Outcomes vary by diagnosis, intensity, duration, and follow-up, and the FTC’s Health Products Compliance Guidance requires substantiation appropriate to the claim rather than aspirational language dressed as fact ¹⁰.

In 2025, the $1.9 million Evoke Wellness settlement addressed Google search ads that impersonated other treatment providers, with a ban on future impersonation of substance-use-disorder clinics ⁵. Brand-adjacent copy, competitor-name bidding paired with ambiguous ad text, and call-center greetings that avoid naming the operating provider now sit inside a live enforcement pattern.

Testimonials, reviews, and the 2024 fake-reviews rule

Reviews and patient stories carry the highest per-word risk in behavioral health creative. The FTC’s final rule banning fake reviews and testimonials took effect October 21, 2024, and prohibits buying, creating, selling, or disseminating fake or false reviews and testimonials, including AI-generated reviews, reviews from people without actual experience, and materially misleading representations of a reviewer’s experience ¹¹. Knowing violations carry civil penalties.

A month later, the FTC’s order against review platform Sitejabber alleged that ratings and reviews were presented as coming from customers who had experienced the products or services when that was not necessarily the case ¹². The action established that intermediaries — review syndicators, reputation-management vendors, AI review tools — sit inside the same liability perimeter as the treatment center itself. Contracts with these vendors need indemnification, provenance logs, and the right to audit whether reviewers actually received care.

Authenticity alone does not clear a testimonial. The FTC’s Health Products Compliance Guidance warns that consumer testimonials and expert endorsements cannot be used to make claims that would be deceptive or unsupported if stated directly ¹⁰. A real patient describing a 30-day transformation implies typical results. If most patients with comparable diagnoses do not achieve that outcome, the testimonial carries a deception risk regardless of the reviewer’s sincerity.

Operationally, the review workflow needs three controls:

• verified proof of care before a story is used,
• disclosure of any material connection or incentive, and
• a claims-review pass that treats the testimonial’s implied outcome the same as any direct efficacy claim in an ad headline.

Professional-ethics overlay for physician-led programs

Programs with physician medical directors carry an additional layer beyond FTC substantiation. The AMA Code of Medical Ethics permits physician advertising when it is truthful and not misleading, and identifies patient testimonials and unsupported claims about skill or outcomes as areas requiring special caution ¹³. Testimonials that do not reflect results generally received by patients with comparable conditions can be deceptive under that standard, and objective claims about experience, competence, and service quality must be factually supportable.

For behavioral health operators, the practical consequence is that ethics review sits alongside legal review, not downstream of it. State medical-board rules and facility licensing requirements add further constraints. Building physician sign-off into the same claims-review workflow that clears FTC risk avoids parallel processes and reduces the chance that ad copy clears legal but exposes the medical director to a board complaint.

Where HIPAA meets pixels, forms, and remarketing audiences

The HHS Office for Civil Rights takes the position that HIPAA applies when information collected through tracking technologies or disclosed to tracking vendors includes protected health information, and that regulated entities must ensure disclosures are permitted by the Privacy Rule and that electronic PHI is secured ². The operational reading for admissions marketing is that a marketing site is not a neutral zone the moment a visitor’s activity begins to relate to their health, healthcare, or payment for healthcare.

That threshold catches more surfaces than most tag-manager setups assume. A contact form asking about diagnosis, insurance, or preferred level of care generates PHI at the point of submission. A chat widget that logs symptom descriptions produces the same result. A page-view pixel firing on a residential-treatment landing page transmits URL and referrer data that, combined with an IP address or ad-platform cookie, can identify an individual seeking care for a specific condition. Remarketing audiences built from those visits and CRM syncs that push inquiry records into ad platforms extend the disclosure chain further.

HIPAA’s marketing rule adds a second constraint. HHS defines marketing as a communication about a product or service that encourages recipients to purchase or use it, and, with limited exceptions, requires written authorization before PHI is used or disclosed for marketing ³. Email nurture sequences, SMS follow-up, and lookalike audience seeding from inquiry lists sit inside that definition unless a specific exception applies.

The audit-ready configuration is per-surface:

• server-side event forwarding that strips identifiers before third-party transmission,
• business associate agreements with every vendor that touches inquiry data,
• form fields limited to what admissions actually needs,
• remarketing audiences built from de-identified events rather than inquiry lists, and
• a documented review of which pixels fire on which templates.

Organizations outside HIPAA’s perimeter do not get a pass — the FTC Act and state privacy laws still apply to consumer health information handled by lead generators, agencies, and technology vendors ⁴.

Phone numbers, call scripts, and account ownership as advertising surfaces

The phone number in a paid-search ad is an advertising claim. So is the greeting the caller hears, the routing logic that decides which facility answers, and the name on the Google Ads account that placed the impression. The FTC’s 2025 complaint against Mercury Marketing alleged violations of the FTC Act, the Opioid Addiction Recovery Fraud Prevention Act, and the Impersonation Rule for search ads that impersonated substance-use-disorder clinics and routed consumers to different providers ⁶. The Evoke Wellness settlement the same month included a ban on impersonating other companies, including other substance-use-disorder treatment clinics ⁵.

Three governance controls follow.

1. First, account ownership must be auditable. Paid-search accounts, tracking numbers, and landing-page domains should sit under the operating provider or a clearly disclosed agent, with written records of who can create ads, add keywords, or change destination URLs. Agency-run accounts that use the agency’s own MCC without a documented audit trail create exposure the clinic ultimately answers for.
2. Second, call-center greetings must name the operating provider within the opening seconds. Generic greetings such as “admissions helpline” or “treatment placement” are the exact pattern the FTC flagged in Evoke Wellness ⁵. Scripts should identify the facility, its location, and its clinical scope before any intake question is asked.
3. Third, call recordings and CRM records fall inside the tracking perimeter described above. Recording consent, retention windows, and downstream use of transcripts for coaching or ad optimization all need HIPAA-authorized workflows, not ad-tech defaults ².

Accessibility as a conversion and legal surface

Inaccessible destinations lose admissions before compliance ever becomes the issue. The Department of Justice’s web guidance identifies alt text, video captions, headings, color contrast, keyboard and mouse navigation, and a way to report accessibility issues as important practices for ADA-relevant web content ¹⁴. Behavioral health audiences include people with visual, motor, cognitive, and hearing disabilities, and family members using assistive technology on behalf of a patient.

Three surfaces produce most of the loss:

• forms that cannot be completed with a keyboard,
• chat widgets that trap screen-reader focus, and
• phone-only conversion paths with no accessible alternative.

Fixing them protects conversion rate on the same landing pages paid media already funds. Legal standards vary by entity and jurisdiction, so accessibility work should be paired with current legal review rather than a generic checklist ¹⁴.

Reach, qualified inquiry, assessment, admission, continuity

Measurement fails in behavioral health when the dashboard stops at cost per lead. The pathway a marketing leader actually owns runs across five stages, and each one produces a different kind of evidence about whether advertising is doing its job.

Reach

The impression and click volume calibrated to the service-line demand mix. It answers whether the media plan is showing up in the moments the earlier layers identified.

Qualified inquiry

Qualified inquiry is the count of contacts that meet a documented threshold: correct geography, appropriate level of care, insurance or self-pay fit, and a clinical scope the facility actually treats. Cost per qualified inquiry, not cost per lead, is the first honest efficiency metric.

Assessment

The count of inquiries that convert into a scheduled and completed clinical evaluation. This is the handoff from marketing to clinical operations, and the drop-off between qualified inquiry and completed assessment usually reveals whether call-center scripts and scheduling logistics are working or leaking.

Admission

Admission is the count of assessments that convert into an active episode of care, weighted by service line and length of stay.

Continuity

Tracks whether the admitted patient remained engaged at 30 and 90 days, transitioned to a lower level of care, or dropped out.

Reporting the five stages separately — with conversion rates between each — is what allows an operator to distinguish a media problem from an intake problem from a clinical fit problem. A campaign with strong reach and weak assessment conversion is not an ad-copy failure.

What the campaign evidence base actually supports

The published evidence on media-based mental health campaigns is more equivocal than most vendor pitches acknowledge. A 2024 systematic review of media campaigns targeting men’s help-seeking for mental-health problems, distress, suicide, or self-harm included 22 studies and found that most campaigns targeting mental health or depression positively influenced help-seeking, though website-based campaign findings were mixed and the overall evidence remained equivocal ¹⁵. The review covered variation in study quality, channels, populations, and outcomes.

Two implications for admissions measurement follow. First, attitude and intention change are legitimate campaign outcomes for awareness-stage media, but they do not substitute for downstream admissions data when justifying a budget. Second, website-based campaign evidence being mixed is a caution against treating landing-page conversion rate as a proxy for real-world help-seeking. The operator move is to instrument both — attitudinal signals for upper-funnel work and admissions-to-continuity data for the paid-inquiry channels — rather than pick one and generalize.